Project III entails an examination of the Florida's Progress Monitoring and Reporting Network (PMRN) database containing multiple measures of multiple reading constructs on 300,000 children attending underachieving schools in the state of Florida. Approximately 124,000 new children enter the database each year in one of the six grade levels in which data are collected: kindergarten through Fifth grade. This extensive database is uniquely equipped to answer questions related to each of the six Center grant goals. It will be used to estimate the percentages of struggling readers who fit alternative classification criteria for learning disability in various domains (e.g., impaired decoding, impaired comprehension, etc.). It also will be used to examine how alternative classification criteria for learning disability (e.g., unexplained low achievement, inadequate growth) are related (i.e., do they classify the same children as learning disabled). We will examine the predictive validity of alternative classification criteria for learning disability, amd environmental variables in relation to alternative classifications (e.g., SES, performance level of the school, etc.) and their predictive validity. For the quantitative genetic studies, we will identify an usually large number (N = 9,000) of identical and fraternal twins within the PMRN sample in order to examine genetic and environmental influence on reading constructs, their relationships to each other, and to their growth over time. We will use the sample within the PMRN to examine differential heritability of reading ability vs. disability defined in alternative ways.